Abstract

Purpose: To examine the predictive validity of the Language Use Inventory (LUI), a parent report of language use by children 18 to 47 months old (O'Neill, 2009).

Method: 348 children whose parents had completed the LUI were re-assessed at 5 to 6 years old with standardized, norm-referenced language measures and parent report of developmental history. The relationship between scores on the LUI and later measures was examined through correlation, binary classification, and ROC curve analysis.

Results: For children aged 24 to 47 months at the time of LUI completion, LUI scores correlated significantly with language measure scores. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) was also calculated for four cut-off scores on the LUI, including –1.64 SD, a score that maximized sensitivity to 81% and specificity to 93%. For children aged 18–23 months at the time of LUI completion, specificity and NPV were high but sensitivity and PPV were lower than desirable.

Conclusions: The results provide initial support for the LUI's predictive validity, particularly for children 24–47 months, and suggest the LUI can serve as an indicator of later language outcomes in referred populations. The results compare favourably to findings for other early child language measures.